Metaphor and the pitfalls of linguistic elicitation
October 20, 2009
This semester at Berkeley I am teaching the year-long graduate field methods course, which I am enjoying tremendously. Apart from having the opportunity to work with a wonderful speaker of Quichua and a group of very intelligent and hard-working students, I have very much appreciated how the experience of helping others get their start with documenting and describing a language has given me the opportunity to reflect on aspects of my own language documentation practices. As in any area in which skills are acquired through praxis, there are aspects of my own practices that I am normally not particularly conscious of, but by watching others find their way and experiment, I’ve become more aware of some of them.
One such aspect concerns the intuitions I have developed about possible pitfalls in elicitation. Among documentary and descriptive linguists, the role of elicitation in a project’s overall methodological toolbox is a source of some controversy, with some linguists, like Tom Payne, recommending a heavy focus on elicitation early on in a documentary/descriptive project, and others, like Bob Dixon, recommending avoiding elicitation until late in a project, after a significant amount of text-based analysis has been carried out. I feel that there are good arguments for both of these elicitation philosophies, but regardless of the approach taken, I believe elicitation involves subtle skills that take time and practice to develop. One especially important aspect of developing elicitation skills, I believe, is acquiring sensitivities to the ways in which the linguistic expectations we have, stemming from the characteristics of our native language(s), can be the source of elicitation difficulties.
One area in which I’ve developed this type of sensitivity, I’ve recently come to realize, is in the use of metaphor. For example, I recently overhead an elicitation session in which a student asked our consultant to translate something like “Dancing gave Fred a headache.” The consultant had some difficulty with translating this sentence, and it dawned on me that I would probably never ask a consultant to translate a sentence like this.
But what kind of sentence is this, and why wouldn’t I ask a consultant to translate it? First, the sentence relies a great deal on metaphor: correctly interpreting the sentence requires interpreting the ‘giving’ expressed in the sentence as a metaphor for causation. And second, it requires understanding (anthropomorphising?) an activity like dancing as capable of ‘giving’ in the first place. Reflecting on this brief interchange and my reaction to it, I realized that over the years, I’ve come to avoid certain metaphors in elicitation. They seem to cause trouble.
But if, as scholars like George Lakoff believe, metaphor is central to human cognition, what exactly am I doing when I am ‘reducing’ my use of metaphor? I think that what I’m really doing is relying on my intuitive sense of what metaphors are cross-linguistically common and which aren’t. For example, I have the sense that the metaphorical use of spatial distances and relations to talk about temporal duration and relations (e.g. ‘a long time’) is very widespread, while the causation-as-giving metaphor discussed above is not. I’m not sure how good my intuition really is, of course, since I’ve only worked with a small number of languages, but the sensitivity I’ve developed seems to be an improvement over having none whatsoever. Of course, what would really be helpful here would a cross-linguistic typological perspective on metaphor, so that we could have a really sound basis for judgements about the use of metaphors in elicitation — but that’s an issue for another post.